46 results about "Auxiliary label" patented technology

The invention discloses a framework for improving the effect of semantic segmentation model based on transfer learning, which comprises the following contents: 1) introducing the transfer learning into the semantic segmentation field, so that the fast semantic segmentation network can improve the segmentation effect of the student model through the teacher model; 2) consistency mapping being proposed to measure the contour and texture information of the teacher-student model, and a consistency loss function being constructed to make the fast semantic segmentation better in detail. 3) using a teacher model and a conditional random field (CRF) model to generate auxiliary label for unlabeled data, and adding data to a training set to improve that generalization ability and segmentation effectof the model. The invention improves the accuracy rate of the fast semantic segmentation model without introducing additional model parameter and reducing the model speed.

The image management device includes a related image selection module and an auxiliary labeling module. The related image selection module selects one or more related images having a certain relationship to a classification labeled image which is assigned a classification label indicating image classification. The auxiliary labeling module assigns an auxiliary label to the related images, the auxiliary label indicating the certain relationship.

A method for producing an analysis tape for fluid samples, in particular body fluids, is proposed. The method comprises providing a laminate tape having a laminate carrier tape and at least one diagnostic functional layer, and cutting the laminate tape in such a way that a diagnostic auxiliary label arises; transferring the diagnostic auxiliary label to a vacuum roller; and transferring the diagnostic auxiliary label to a carrier tape. In general, at least one vacuum roller is used for the transfer of the diagnostic auxiliary label to the carrier tape. The cutting of the diagnostic functional layer is effected in such a way that a free end of the diagnostic functional layer is lifted off from the laminate carrier tape by guiding the laminate carrier tape around a cutting edge with a small radius, wherein the diagnostic functional layer is cut at a predetermined distance from the lifted off free end by means of a cutting device.

The invention discloses an auxiliary label stripping jig which is applicable to the tape label comprising a substrate layer and multiple labels attached to the substrate layer. The jig comprises a bottom plate, a bearing structure, an import block, a stripping block and a top plate. The bearing structure fixes the tape label and the label can be pulled in a gap in the structure; the import block strips the substrate with certain width from the label; and an inner collection gap formed by the stripping block and the bearing structure expands the stripping degree between the substrate layer and the label in the tape label pulling process. In addition, the jig can be adjusted through an adjusting/positioning device of the bearing structure so as to adapt to the tape belts of different widths. According to the invention, the trouble of manually stripping label can be avoided, and the label and the substrate layer are directly stripped to certain width so that the label can be fetched conveniently by the operator.

The invention relates to the power supply equipment operation and maintenance technology, in particular to an operation and maintenance auxiliary tag generation method and system, and the method comprises the following steps: A) importing an equipment account and historical information; B) collecting and returning equipment information and label setting when an operation and maintenance worker carries out on-site maintenance; c) establishing and training a neural network model; d) after follow-up on-site maintenance of similar equipment, operation and maintenance personnel upload equipment information and labels, update the equipment information and substitute the updated equipment information into the neural network model in the step C) to obtain model labels, and counting model correct rate in a period; and E) if the model accuracy reaches a set threshold value in the statistical period, generating labels for all similar devices. The method has the advantages that through generationand accumulation of the operation and maintenance labels, operation and maintenance-oriented information guidance and experience reference can be provided for subsequent same equipment and similar equipment, operation and maintenance efficiency is improved in a targeted assisting mode, and the problem that operation and maintenance of a power grid are heavy at present is solved.

The invention discloses a bank card auxiliary system based on Beidou positioning, comprising an auxiliary label arranged on a bank card for storing bank card number, transmitting a request of bank card position information and bank card number to a Beidou navigation system; a data processing module for receiving the bank card position information and user terminal position information, which are transmitted by the Beidou navigation system, and calculating relative distance between the bank card position information and the user terminal position information, and calculating a distance difference value of an alarm distance threshold value and the relative distance, and transmitting the relative distance and the distance difference value to the Beidou navigation system, a communication module which is arranged on a user terminal for transmitting the request of the user terminal position information to the Beidou navigation system and receiving communication data. The bank card auxiliary system based on Beidou positioning of the invention can be used for accurately and fast obtaining position information of the bank card and the user terminal position information, and the relative distance of the bank card and the user terminal.

The invention discloses a thin-walled cylinder hot melt adhesive labeling machine. The thin-walled cylinder hot melt glue labeling machine comprises a bottle carrying and transferring table, a bottle feeding device, a glue spraying device, a split floating label bin, a glue scrapping device, a glue melting box and a bottle discharging device, wherein the bottle carrying and transferring table are installed in the center of the table surface of a rack, and the bottle feeding device, the glue spraying device, the split floating label bin, the glue scrapping device, the glue melting box and the bottle discharging device are located on the table surface of the rack and distributed on the periphery of the bottle carrying and transferring table. An adjustable label box is installed at the front end of a base plate in the split floating label bin, an auxiliary label feeding device is installed at the rear end of the base plate, elastic label pushing devices are symmetrically installed on the left side and the right side of the base plate correspondingly, and a rolling label feeding chain way is installed in the middle of the base plate. According to the thin-walled cylinder hot melt glue labeling machine, modularized design and a mechanical automatic production line of the composition structure are achieved, whole cylinder body labeling of labels with large diameters and widths of thin-walled paper cylinders, thin-walled metal cylinders and the like can be achieved, the replacing frequency of the labels can be greatly reduced through the label bin with a high capacity, operation of an operator is facilitated, production efficiency is greatly improved, the labeling quality and labeling precision are further improved, and production cost is maximally reduced.

The invention provides a data sorting process dynamic implementation method for teaching demonstration. The data sorting process dynamic implementation method adopts a data number edit module, a sorting data random generation module, an initial data label generation module, an initial sorting data display module, a data storage and reading module, a data sorting module, an auxiliary label generation module, a forward comparison data label attribute modification module, a data exchange process dynamic display module, a sorted data label attribute modification module and a sorted data result display module. When the data sorting process dynamic implementation method is used for implementing explanation of a programming course data sorting algorithm, the specific data sorting process can be demonstrated dynamically, the data exchange sorting process is demonstrated dynamically in the sorting process according to presorting data generated randomly, and the sorted results are displayed after sorting. By applying the data sorting process dynamic implementation method, data sorting theoretical process can be demonstrated for the students vividly, the learning efficiency of students is increased, and good teaching effect is achieved.

The invention discloses a label structure and an attaching method thereof. The label structure, which can be attached in a label area on an article, comprises a main label and an auxiliary label, and the main label is printed with main information, and is provided with a transparent window. The auxiliary label is printed with auxiliary information, wherein the main label is attached in the label area, and the auxiliary label is clamped between the main label and the article, and is opposite to the transparent window, so that the transparent window can display the auxiliary information of the auxiliary label. The method includes the following steps: the main label, which is printed with the main information and provided with the transparent window and is waterproof, is provided; the auxiliary label, which is printed with the auxiliary information, is provided; one surface of the auxiliary label, which is printed with the auxiliary information, is attached on the main label, and is opposite to the transparent window, so that the transparent window can display the auxiliary information of the auxiliary label; the main label is attached in the label area of the article, so that the auxiliary label is clamped between the main label and the article, and thereby the probability of contact between outside liquid or foreign matters and the auxiliary label is decreased because of the waterproofness of the main label.

The present invention relates to child-resistant cap and prescription pill bottle for dispensing prescription medications, such as tablets and capsules. The present inventions of push-down-and-turn child-resistant caps relate to two novel closure mechanisms and one improved mechanism using tapered detent in the camming latch that lead to the locking notch for smoother operation, as well as cap with ergonomic grip. All three mechanisms offer easier operation with a simple resilient disc member. The present invention of prescription pill bottle is a plurality of elliptical cylindrical shaped bottle which offers ergonomic grip compare to the circular cylindrical shape currently used. In addition, the bottles offer a relatively planar surfaces for prescription and auxiliary labels, which eliminate the action of rotating the bottle back-and-forth to read the labels.

A method of auxiliary label detection includes: at an imaging controller, obtaining (i) an image of a support structure and (ii) a plurality of primary label indicators each defining a position of a primary label within the image; at the imaging controller, selecting a detection region from the image based on a reference indicator; at the imaging controller, detecting an auxiliary label within the detection region based on predefined visual attributes of the auxiliary label; at the imaging controller, selecting one of the primary label indicators based on proximity between the auxiliary label and the selected primary label indicator; updating the selected primary label indicator with an identifier of the auxiliary label; and storing the updated primary label indicator in a memory.

The invention provides a data message packaging method and device. The method comprises the following steps that based on a destination address of a data message needing to be forwarded, an exit tableitem is searched out, wherein the exit table item comprises an exit auxiliary label, an intermediate node label, an exit node label and a loopback port; the exit auxiliary label, the intermediate node label and the exit node label of the exit table item are packaged for the data message, a loopback port outer layer Ethernet head is packaged according to the loopback port, and an MPLS data messagewith the exit auxiliary label is sent through the loopback port; the MPLS data message with the exit auxiliary label is received through the loopback port; the exit auxiliary label is stripped when an entrance table item is searched out according to the exit auxiliary label; and an entrance node label and an intermediate node label of the entrance table item are packaged, an exit port outer layerEthernet head is packaged according to an exit port, on a forwarding path, of an entrance node in the entrance table item, and the data message is sent through the exit port. The multi-layer label packaging cost can be effectively reduced.

The embodiment of the invention discloses a buffering method and device during paper classification, equipment and a storage medium, and belongs to the technical field of text information processing.The method comprises the steps of acquiring a to-be-classified text set and a classification target set; obtaining a classification label tree of the classification target set and a classification label set of each to-be-classified text; taking label elements in the classification label set as query conditions, and obtaining an auxiliary label set from the classification label tree; based on the label elements in the auxiliary label set and a preset screening condition, screening out a historical classification model meeting the preset condition from historical classification models of a classification target set; and selecting a final classification model from an initial test classification model set or newly establishing a final classification model, and classifying the to-be-classifiedtexts in the classification text set based on the final classification model. According to the method, the text classification efficiency of a user is improved, the availability of the historical classification model is maintained, and a better classification service is provided for the user.

The embodiment of the invention discloses a live broadcast type determination method and device, equipment and a storage medium. The method comprises the steps of determining an auxiliary label of a target live broadcast room according to live broadcast evaluation information of the target live broadcast room and/or a calibration label of a calibration live broadcast room; determining a complete label of the target live broadcast room according to the original label and the auxiliary label of the target live broadcast room; and determining the live broadcast type of the target live broadcast room according to the complete label. By adopting the technical scheme provided by the embodiment of the invention, the problem that live broadcast types of a plurality of network live broadcast roomscannot be clear, and then accurate media information release cannot be carried out according to the live broadcast type of the live broadcast room is solved. According to the embodiment of the invention, the auxiliary tag is added to the live broadcast room and the original tag of the live broadcast room is combined to form the complete live broadcast tag of the live broadcast room, so that the live broadcast type of the live broadcast room can be accurately determined according to the complete live broadcast tag to reasonably divide the type of the live broadcast room.

An auxiliary prescription label is provided for print-impaired individuals, such as blind and vision-impaired individuals. The auxiliary label is affixed to a prescription receptacle and does not permanently obscure the primary prescription label. The auxiliary label includes a book attached to a clip. The book is formed by a strip of printed material. The strip is folded to form the book and at least one stack of apertures is provided therethrough. The clip includes a base having at least one ear extending therefrom. The ear(s) are bent relative to the base to pass the ear(s) through the aperture(s), and the ear(s) are thereafter secured to the base. An adhesive secures the auxiliary label to the prescription receptacle.

The invention discloses an auxiliary data labeling method capable of achieving online learning. The method comprises the steps of first-time model training, model marking data using, marking data manual correction and model re-training optimization. According to the auxiliary data labeling method disclosed by the invention, model training can be completed under the condition of few initial test data, and the model is used for auxiliary data labeling; most importantly, the online learning is realized by training the model again by using the data marked by the auxiliary label each time.The performance of the model is further improved, and the accuracy of target detection of the model is improved, so that the accuracy of auxiliary data annotation is improved, and the consumption of time and labor cost in a repeated manual data annotation process is greatly reduced.

The invention discloses a multi-angle tilting induction auxiliary device comprising an inductive auxiliary label body. A an inductive device is fixedly connected to the top of the inductive auxiliarylabel body and includes a tilting sensing base; an inductive prompt light is fixedly installed at the top of the tilting sensing base; and a protective sensing control cover is fixedly connected to the top of the tilting sensing base. An inductive tilt control ball moves towards an inclination direction and enters a tilt control opening in a corresponding direction; and when the inductive tilt control ball enters the inside of the tilt control opening, a lighting control switch inside the tilt control opening moves down and presses under the effect of the gravity and thus the corresponding inductive prompt light turns on, so that the transport angle of the cargo is adjusted conveniently and thus the balance is kept in the cargo conveying process to realize the multi-angle tilt sensing. Therefore, a problem that the existing induction auxiliary device can not carry out multi-angle simultaneous sensing and thus the cargo protection is affected is solved.

A label producing apparatus has a pressure roller drive shaft for feeding a cover film supplied from a cartridge storing a cover film capable of supply, acquires the arrangement information by accessing a database storing and holding arrangement information of the article to be arranged, and has a print head and a cutter for producing an auxiliary label by forming the acquired arrangement information for the cover film fed by the pressure roller drive shaft.

The invention relates to the technical field of prediction models, and discloses an interview entity identification model training method and device and an interview information entity extraction method and device. According to the method, standard label prediction is carried out on first interview sample data through a direct prediction module in a preset identification model; standard label distribution and interview coding vectors are obtained; auxiliary label prediction is carried out on the first interview sample data according to the coding vector through each auxiliary prediction module to obtain auxiliary label distribution output by each auxiliary prediction module; a total loss value of a preset identification model is determined according to the auxiliary label distribution and the standard label distribution; and when the total loss value does not reach the preset convergence condition, a first initial parameter of the iterative preset identification model is updated until the total loss value reaches the preset convergence condition, and the converged preset identification model is recorded as an interview entity identification model. According to the invention, the model training efficiency and the model identification accuracy are improved.

The invention discloses a positioning method for non-reflector laser navigation by adopting an auxiliary. The method comprises the following steps: a navigation path of a non-reflector laser navigation system is defined as a 0 type path and a I type path which is of different aberration with the ground is arranged in a section with precision requirement; a visual sensor mounted on a mobile platform of the non-reflector laser navigation system photographs the I type path at a certain frequency to acquire an offset distance from a central line of the visual sensor to a central line of the I type path, that is, the transverse deviation h of the mobile platform; and when the mobile platform travels on the I type path, the mobile platform acquires a deflection angle alpha of the mobile platform and a real-time distance I of longitudinal movement through the non-reflector laser navigation system, and calculates accurate real-time posture information according to h, alpha and l so as to correct the posture of the mobile platform in real time. According to the positioning method disclosed by the invention, in a station which is relatively high in precision demand, stable coordinate information is formed by adding a local auxiliary label and combining with a non-reflector laser technology, so that the positioning precision is greatly improved and the precision is stably kept at +/-10mm.

The invention belongs to the technical field of basketball processing, and particularly relates to a noctilucent basketball and a coating method thereof. The noctilucent basketball comprises a basketball skin, the basketball skin is sequentially provided with a middle tire, a winding yarn and an inner container from outside to inside, the middle tire, the winding yarn and the inner container are bonded and fixed, and an air nozzle is formed in the basketball skin; a plurality of middle tire groove ribs are uniformly distributed on the basketball skin; 1000-mesh noctilucent powder and clear oilare proportioned, then a screen printing plate and a copper plate in a corresponding shape are opened and pressed on the surface of a basketball, so that the manufactured basketball main label and the manufactured basketball auxiliary label have a noctilucent effect, the proportion of the noctilucent powder can be increased, the light-emitting time can be correspondingly prolonged, and the brightness can be increased; by adoptingscreen printing plate multi-time coloring and imprinting copper plates with corresponding shapes of the main label and the auxiliary label on the surface of the basketball, the noctilucent position is lower than the surface layer of the basketball leather, and the situations that the basketball is abraded and fluorescent powder is easy to fall off in the using process are avoided.